The Transistor
Why This Matters
The vacuum tube proved that electrons can be controlled to amplify and switch signals, but tubes are large, fragile, power-hungry, and short-lived. The transistor does everything a vacuum tube does in a package the size of a pea, using a fraction of the power and lasting decades. It is the single most important invention in the history of technology. Without transistors, there are no computers, no digital communications, no modern civilization. This article teaches you how they work and how to build the simplest ones from raw materials.
What You Need
For understanding and using transistors:
- Salvaged transistors (any NPN type: 2N2222, 2N3904, BC547, or equivalent)
- Resistors: assorted values from 100 ohms to 1M ohm
- Capacitors: 0.01 uF to 100 uF
- LEDs or small lamps for visual output
- Multimeter (voltage, current, resistance)
- Breadboard or perfboard
- 5-12V DC power supply or batteries
- Soldering iron and solder
For fabricating primitive transistors:
- Germanium or silicon crystal (salvaged from diodes or raw mineral)
- Fine phosphor bronze or tungsten wire, 0.05-0.1 mm (for point contacts)
- Micromanipulator or very steady hands and magnification
- Etching chemicals: hydrochloric acid, nitric acid
- Indium metal (for alloy junction method)
- Heat source capable of 160-400 C (for alloying)
Semiconductor Physics Refresher
To understand transistors, you must understand what makes semiconductors special. They are neither conductors nor insulators --- they are in between, and their conductivity can be precisely controlled by adding impurities.
The Band Gap
In any solid material, electrons exist in energy bands:
- Valence band: Electrons are bound to atoms, participating in chemical bonds. These electrons do not conduct.
- Conduction band: Electrons are free to move through the material. These electrons carry current.
- Band gap: The energy difference between the top of the valence band and the bottom of the conduction band.
| Material | Band Gap (eV) | Behavior |
|---|---|---|
| Copper | 0 (bands overlap) | Conductor --- electrons flow freely |
| Germanium | 0.67 | Semiconductor --- some thermal excitation |
| Silicon | 1.12 | Semiconductor --- less thermal excitation |
| Diamond | 5.5 | Insulator --- virtually no conduction |
| Glass | ~9 | Insulator |
At room temperature, silicon has very few electrons in the conduction band. But by adding impurities (doping), you can put electrons exactly where you want them.
N-Type and P-Type
N-Type (negative carriers):
- Add atoms with 5 valence electrons (phosphorus, arsenic) to silicon (4 valence electrons)
- The extra electron from each dopant atom is free to conduct
- The material has an excess of negative charge carriers (electrons)
P-Type (positive carriers):
- Add atoms with 3 valence electrons (boron, gallium, indium) to silicon
- Each dopant atom creates a โholeโ --- a missing electron that acts like a positive charge carrier
- Holes move through the material as neighboring electrons fill them, creating new holes
Tip
Think of holes like an empty seat in a crowded theater. When someone shifts to fill the empty seat, they create a new empty seat where they were. The โempty seatโ appears to move through the theater, even though it is really people moving the other way. Holes move through a semiconductor the same way.
The PN Junction
When P-type and N-type materials meet, something remarkable happens at the boundary.
The Depletion Zone
At the junction:
- Free electrons from the N-side diffuse into the P-side
- Holes from the P-side diffuse into the N-side
- Where they meet, electrons fill holes, creating a thin region with no free carriers
- This region is the depletion zone --- it acts as an insulating barrier
- The depletion zone creates a built-in voltage (about 0.3V for germanium, 0.6V for silicon)
Forward and Reverse Bias
Forward bias (positive to P-side, negative to N-side):
- The external voltage pushes against the built-in voltage
- The depletion zone shrinks
- Above the threshold voltage (0.6V for silicon), current flows freely
- The junction conducts
Reverse bias (positive to N-side, negative to P-side):
- The external voltage reinforces the built-in voltage
- The depletion zone widens
- Almost no current flows (just a tiny leakage current)
- The junction blocks
This is a diode --- the solid-state equivalent of the vacuum tube diode, but smaller, cheaper, and more robust.
The Bipolar Junction Transistor (BJT)
A transistor is two PN junctions back-to-back, sharing a thin middle layer. The three layers are called Emitter, Base, and Collector.
NPN Transistor
The most common type: N-type emitter, P-type base, N-type collector.
Collector (N)
|
[N material]
========================
[P material] <--- Base (very thin, ~1 micrometer)
========================
[N material]
|
Emitter (N)
How it works:
-
With no base current, both junctions are effectively off. The collector-base junction is reverse-biased by the supply voltage, so no current flows from collector to emitter. The transistor is OFF.
-
Apply a small forward bias voltage to the base-emitter junction (above 0.6V for silicon). Current flows into the base.
-
Because the base is extremely thin, most of the electrons injected from the emitter do not recombine with holes in the base. Instead, they are swept across into the collector by the electric field of the reverse-biased collector-base junction.
-
A small base current (microamps) controls a much larger collector current (milliamps). The ratio is the current gain (beta or hFE), typically 50-300.
Collector current (Ic) = Beta x Base current (Ib)
Example: Beta = 100, Ib = 0.1 mA
Ic = 100 x 0.1 = 10 mA
PNP Transistor
Same principle, reversed polarities. The emitter is P-type, base is N-type, collector is P-type. Current flows in the opposite direction. PNP transistors are used in complementary circuits alongside NPN types.
Tip
When troubleshooting transistor circuits, remember the key voltages for silicon transistors: the base-emitter voltage is always approximately 0.6-0.7V when the transistor is conducting. If you measure significantly more or less, the transistor is likely damaged or the circuit has a fault.
The Transistor as a Switch
For digital electronics and computing, the transistor operates in only two states: fully ON (saturation) or fully OFF (cutoff). No in-between.
Switching Circuit
+Vcc (5V)
|
[Load: LED + 330 ohm resistor]
|
Collector
|
[NPN Transistor]
|
Emitter
|
GND
Base ---[10k resistor]--- Control Signal (0V or 5V)
When the control signal is 0V: No base current flows. The transistor is OFF. No current through the LED. Output is HIGH (approximately Vcc).
When the control signal is 5V: Base current flows through the 10k resistor: Ib = (5V - 0.6V) / 10k = 0.44 mA. With beta = 100, this can control Ic = 44 mA. The LED only needs ~15 mA, so the transistor is fully saturated (ON). The collector voltage drops to about 0.2V (approximately ground). Output is LOW.
Key insight for computing: The output is the inverse of the input. Input HIGH gives output LOW, and vice versa. This is a NOT gate --- the simplest logic operation, built with a single transistor.
Switching Speed
| Parameter | Typical Value | Significance |
|---|---|---|
| Turn-on time | 10-100 nanoseconds | How fast the transistor switches from OFF to ON |
| Turn-off time | 50-500 nanoseconds | OFF takes longer due to stored charge in the base |
| Propagation delay | 5-50 nanoseconds | Time for input change to appear at output |
| Maximum frequency | 10-300 MHz | Upper limit for reliable switching |
Even the slowest discrete transistors switch millions of times per second. Compare this to a vacuum tube, which might manage a few megahertz with difficulty. This speed advantage is what makes digital computing practical.
The Transistor as an Amplifier
For analog applications (audio, radio, sensors), the transistor operates in its active region --- between cutoff and saturation, where small input changes produce proportional large output changes.
Common Emitter Amplifier
The most widely used configuration, analogous to the vacuum tube common cathode amplifier:
+Vcc (12V)
|
[Rc: 4.7k] (Collector resistor)
|
+----+----> Output
|
Collector
|
[NPN Transistor]
|
Emitter
|
[Re: 1k] (Emitter resistor)
|
GND
Voltage divider bias:
+Vcc---[R1: 47k]---+---[R2: 10k]---GND
|
Base
Input --[Coupling Cap: 10uF]-- Base
Design procedure:
-
Choose the operating point: Set collector current at about half of Vcc/Rc. For Vcc = 12V and Rc = 4.7k, maximum Ic = 12/4.7k = 2.6 mA. Aim for Ic = 1 mA.
-
Calculate emitter voltage: Ve = Ic x Re = 1 mA x 1k = 1V.
-
Calculate base voltage: Vb = Ve + 0.6V = 1.6V.
-
Set voltage divider: Vb = Vcc x R2/(R1+R2). With R1 = 47k and R2 = 10k: Vb = 12 x 10/57 = 2.1V. Close enough (the emitter resistor provides feedback that stabilizes the operating point).
-
Voltage gain: Av = Rc / Re = 4.7k / 1k = 4.7 (without bypass capacitor). With a bypass capacitor across Re, gain increases to approximately Rc x gm, which could be 100 or more.
Tip
The emitter resistor Re stabilizes the circuit against temperature changes and variations in transistor beta. Without it, the operating point drifts as the transistor warms up, potentially destroying it through thermal runaway. Always include Re, even if you bypass it with a capacitor for AC gain.
Fabricating Primitive Transistors
The Point-Contact Transistor
This is what Bardeen and Brattain built in 1947 --- the first transistor ever made. It is the easiest type to replicate.
Materials:
- A slab of N-type germanium (salvage from old germanium diodes, or use a natural germanium crystal treated with arsenic vapor)
- Two fine wires (phosphor bronze or gold, 0.05-0.1 mm diameter), called โcatโs whiskersโ
- A base contact on the bottom of the germanium (soldered or pressure contact)
Construction:
- Polish the germanium surface flat and clean with acid etch (dilute HCl)
- Mount the germanium on a metal base plate (this is the base contact)
- Position two catโs whisker wires to touch the germanium surface, about 0.05-0.1 mm apart (this is the critical dimension --- too far apart and there is no transistor action)
- Apply a brief pulse of current (forming pulse) to one whisker: this creates a tiny P-type region under the contact by electrical forming
- That whisker becomes the emitter, the other becomes the collector
Testing:
- Apply a small voltage between base and emitter (forward bias, about 0.3V for germanium)
- Measure collector current
- You should see current gain (alpha) of 1-3 (modest, but it proves the concept)
Warning
Point-contact transistors are extremely fragile. The whisker contacts can shift with vibration, changing or destroying the transistor action. They are proof-of-concept devices, not practical for circuits. The alloy junction method is far more reliable.
The Alloy Junction Transistor
A more robust approach:
- Start with a thin wafer of N-type germanium (0.5-1 mm thick)
- Place small dots of indium (a P-type dopant) on opposite faces
- Heat to about 160 C (indium melts at 157 C)
- The molten indium dissolves into the germanium, creating P-type regions
- When cooled, you have a PNP junction: P (indium-doped) - N (original germanium) - P (indium-doped)
- Solder leads to each indium dot and to the germanium edge
This produces a PNP transistor with current gains of 10-50 and reasonable reliability.
Common Mistakes
| Mistake | Why Itโs Dangerous | What to Do Instead |
|---|---|---|
| Exceeding maximum collector current | Transistor overheats, junction damage, permanent failure | Check datasheet Ic(max), design for 50-80% of limit |
| No base resistor | Excessive base current destroys base-emitter junction | Always calculate and include a base limiting resistor |
| Thermal runaway in amplifiers | Rising temperature increases current, which increases temperature, until destruction | Use emitter degeneration resistor, proper heat sinking |
| Forgetting flyback diode with inductive loads | When transistor switches off relay/motor, voltage spike exceeds breakdown | Always put a reverse diode across inductive loads |
| Applying reverse voltage to base-emitter | Junction breaks down at only 5-7V reverse | Protect with clamp diode or resistor to limit reverse voltage |
| Static discharge handling | Gate oxide (FET) or thin junction destroyed by static | Handle with grounded wrist strap, touch ground before touching components |
| Wrong pin identification | Collector and emitter swapped, circuit does not work or transistor is damaged | Always verify pinout with datasheet or multimeter diode test |
| Using transistor above frequency limit | Gain drops to unity, switching becomes unreliable | Check fT (transition frequency), operate well below it |
Whatโs Next
With the transistor, you have a tiny, efficient, reliable switch and amplifier. The next critical step is combining transistors into logic gates:
- Boolean Logic and Gates --- learn how to combine transistors into AND, OR, NOT, NAND, and NOR gates that perform logical operations, forming the foundation of all digital computing
Quick Reference Card
The Transistor --- At a Glance
What it does: Controls large current with small current (amplification) or switches fully on/off (digital logic)
Types: NPN (most common), PNP (complementary), point-contact (historical), alloy junction (early practical)
Key voltage: Base-emitter = 0.6V for silicon, 0.3V for germanium (when conducting)
Current gain (beta): Typically 50-300 for discrete BJTs
Switching states: Cutoff (OFF, no current), Saturation (ON, full current, Vce approximately 0.2V)
Common emitter gain: Av = Rc / Re (without bypass capacitor)
Switching speed: 10-500 nanoseconds (millions of times faster than mechanical switches)
Thermal protection: Always use emitter resistor, heat sink for power transistors, stay below 80% of maximum ratings
Fabrication: Point-contact (2 whiskers on germanium, 0.05 mm apart) or alloy junction (indium dots melted into germanium at 160 C)
NOT gate: Single NPN transistor with collector resistor inverts the input signal --- the foundation of digital logic
Critical rule: Never exceed maximum ratings for voltage, current, or power dissipation